Active engineering prototype

AiME — AI-assisted Mechanical Design

Conversational mechanical CAD for parts, fixtures, and enclosures, combining parametric models with STEP references, verified PCBA intake, and measured assembly geometry.

Challenge

Connect design intent to editable mechanical models and make board-to-enclosure integration easier to inspect and measure.

Constraints

Mechanical conclusions must come from the actual geometry and declared references. A rendered assembly or collision check cannot establish manufacturing suitability or physical fit.

System

A desktop workspace connects conversational design to parametric Python models and a geometry-processing sidecar. STEP references and verified PCBA packages compose into a working assembly with measurable distance and intersection results.

Magneum’s contribution

Developed the conversational part-editing workflow, validated reference intake, assembly controls, and exact geometry checks for mechanical CAD work.

Verification

A recorded live workflow proposed and applied a parametric part edit through the operator interface. Independent checks cover STEP and PCBA intake, composed assembly rendering, and pairwise geometry evaluation.

Outcome

AiME supports an evidence-based MCAD workflow from bounded conversational edits to reference-based mechanical integration. Enclosure work has progressed to a clamshell design awaiting its first physical fit test.

Limitations

Physical enclosure fit and manufacturing readiness remain unproven. Some component geometry is based on review models pending vendor STEP files; general app-owned STEP/STL export and persistent assemblies remain open.

Related challenge

Need to make a complex system testable?

Start a conversation